transformWeka.java

transformWeka.java

   /**
     * load training data and set feature generators
     */
    public void transform() {
        try {
            trainData = loadDataset(TRAIN_DATA);
            saveArff(trainData, TRAIN_ARFF_ARFF);
            /**
             * create the filter and set the attribute to be transformed from text into a feature vector (the last one)
             */
            StringToWordVector filter = new StringToWordVector();
            filter.setAttributeIndices("last");
            /**
             * Add ngram tokenizer to filter with min and max length set to 1
             */
            NGramTokenizer tokenizer = new NGramTokenizer();
            tokenizer.setNGramMinSize(1);
            tokenizer.setNGramMaxSize(1);
            /**
             * Tokenize based on delimiter
             */
            tokenizer.setDelimiters("\\W");
            filter.setTokenizer(tokenizer);
            /**
             * To lowercase converting
             */
            filter.setLowerCaseTokens(true);
            /**
             *  Set filter to classifier
             */
            classifier.setFilter(filter);
        } catch (Exception e) {
            LOGGER.warning(e.getMessage());
        }
    }